FISHERY BULLETIN: VOL. 86, NO. 2 



barrier to settlement to neustonic juvenile red 

 hake, Urophycis chuss, (Steiner et al. 1982; 

 Steiner and 011a 1985). 



In contrast, juvenile sablefish apparently re- 

 main in the water column off Oregon following 

 the transition to upwelling conditions. Small ju- 

 veniles (<50 mm) have been captured as far as 

 250 km offshore in the spring (Kendall and Clark 

 1982a), and specimens up to 250 mm have been 

 collected from the surface waters in late summer 

 with a purse seine (Brodeur and Pearcy 1986). 

 The very rapid growth rates of young juveniles, 

 reaching 2 mm/day (Boehlert and Yoklavich 

 1985; Shenker and 011a 1986), undoubtedly re- 

 sulted in their increasing ability to avoid the 

 towed neuston nets used in this study. 



The most abundant larvae occurring after the 

 onset of upwelling were northern anchovy and 

 rockfish. As observed in previous studies 

 (Richardson 1973, 1980; Richardson et al. 1980), 

 these larvae were primarily found offshore. The 

 occurrence of these species, along with a high 

 abundance of vertically migrating blue lantern- 

 fish, on the periphery of the Columbia River 

 plume 60-90 km offshore on only one cruise, fur- 

 ther illustrates the patchy nature of neustonic 

 distributions. 



Although the fish fauna in 1984 was generally 

 characterized by discrete temporal and/or spatial 

 limitations, the occurrence of Dungeness crab 

 megalopae transcended these limits through the 

 4-mo study. Megalopae were the most abundant 

 organisms collected throughout the survey, but 

 their abundance varied widely between adjacent 

 stations, with occasional very dense patches. Sim- 

 ilar patchy distributions of megalopae were ob- 

 served off British Columbia by Booth et al. (1985), 

 who measured horizontal patch dimensions of 

 2-4 km. Several dense swarms of megalopae were 

 observed in Bodega Bay, CA, during 1985 

 (Shenker and Botsford^). These patches were 

 long, sinuous aggregations extending 10-20 m 

 along the surface, approximately circular in cross 

 section and about 1 m in diameter. Densities were 

 visually estimated to be on the order of thousands 

 of megalopae per m^. 



The temporal occcurrence of crab larvae in the 

 plankton spans two distinctly different oceanic 

 regimes, and the larvae are potentially trans- 

 ported long distances by the seasonal currents. 



^Shenker, J. M., and L. W. Botsford, University of California, 

 Bodega Marine Laboratory, P.O. Box 247, Bodega Bay, CA 

 94923, unpubl. data. 



Larvae typically hatch in mid-winter, pass 

 through 5 zoeal stages in approximately 90 days, 

 and then a month-long megalops stage before set- 

 tling to the bottom (Reilly 1983a). After hatching, 

 zoeae are released into the northerly flowing 

 Davidson Current. Despite a general onshore 

 component of flow of the Davidson Current, older 

 zoeal stages have been found progressively far- 

 ther offshore (Lough 1976; Reilly 1983a). 



About the time of the spring transition in the 

 alongshore currents from a northerly to southerly 

 direction (Huyer et al. 1975), zoeae metamor- 

 phose into megalopae. To survive, these megalo- 

 pae must be transported back toward the shore, 

 and settle to the bottom in depths shallower than 

 25 m (Reilly 1983b). Larvae have been found at 

 least 100 km from shore (Lough 1976), although 

 it is unclear if these larvae make it back to shore, 

 or are simply lost from the population. 



Again, an apparent anomaly exists between the 

 directions of movement of megalopae and surface 

 waters, where the Ekman layer moves offshore in 

 response to upwelling winds. These discrepancies 

 may be explained by mesoscale mixing processes, 

 as cited earlier. Alternatively, the diel vertical 

 movements of the larvae can move them into dif- 

 ferent water masses with different directions of 

 zonal movement. 



The data from this study and previous research 

 on vertical migration indicate that at least the 

 early zoeal stages and megalopae move to the sur- 

 face during twilight, and below the surface dur- 

 ing the day (Reilly 1983a). Surface abundance 

 estimates of megalopae obtained in the 27-h sam- 

 pling on 8-9 June (Fig. 8), and by Booth et al. 

 (1985), decreased during the middle of the night. 

 This movement away from the surface is an ex- 

 ample of "midnight scattering", perhaps result- 

 ing from the lack of a light cue to orient plank- 

 tonic organisms to the surface (Owen 1981). If 

 midnight scattering is typical for megalopae, 

 abundance estimates made by sampling along a 

 transect throughout the night (Figs. 7, 9) proba- 

 bly underestimate the actual abundance of mega- 

 lopae utilizing the surface habitat. Megalopae 

 have been collected as deep as 50-70 m during 

 the day (Booth et al. 1985; Shenker and Botsford 

 fn. 6). These observations correlate with Jacoby's 

 (1982) laboratory demonstration that megalopae 

 are positively phototactic to dim light, but avoid 

 bright light. 



The present study indicates that the phototac- 

 tic response of megalopae may assist their return 

 to shore in several ways. Downward movement 



314 



